IBM Cloud Kubernetes Service

A container is an executable software unit in which application code is packaged — together with libraries and dependencies — so it can run anywhere on the desktop, traditional IT or in the cloud. Containers exploit a form of OS virtualization that lets applications share the OS by isolating processes and controlling access to CPU, memory and processes.

As containers proliferated, operations teams needed to schedule and automate container deployment, networking, scalability and availability. Kubernetes became the most widely adopted for its functionality, its ecosystem of open source-supporting tools and its portability across leading cloud providers, some of which offer fully managed Kubernetes services.

Clusters comprise nodes. Each node represents 1 compute host. Worker nodes in a cluster deploy, run and manage containerized apps. Pods share compute resources and network and are key to scaling. If a container in a pod gets too much traffic, Kubernetes replicates the pod. Deployments control the creation and state of the containerized app and keep it running.

As the number of containers in a cluster grows, the possible connection paths between them grow exponentially, making configuration and management complex. Istio on IBM Cloud, an open source service mesh layer for Kubernetes clusters, adds a sidecar container to each Kubernetes cluster. A sidecar configures, monitors and manages interactions between other containers.

Knative, an open source platform, sits on top of Kubernetes and provides two vital benefits for cloud-native development. It’s an easy access method to serverless computing and a way to build a container once and run it as a software service or serverless function. Knative transparently handles things like generating configuration files and writing CI/CD scripts.

Docker provides the containerization piece, enabling developers to easily package applications into small, isolated containers using the command line. When demand surges, Kubernetes provides orchestration of Docker containers, scheduling and automatically deploying them across IT environments to ensure high availability.

Ingress in Kubernetes is an API object that provides routing rules to manage external users' access to the services in a Kubernetes cluster. With Ingress, you can easily set up rules for routing traffic without creating a bunch of load balancers or exposing each service on the node. This makes it the best option to use in production environments.

Docker Swarm is deployed using Docker Engine and is readily available in your environment. Swarm is easier to start with and may be more ideal for smaller workloads. 

Kubernetes is more powerful, customizable and flexible, but has a steeper learning curve. Running Kubernetes with a managed service simplifies open source management responsibilities and allows you to focus on building applications.

Kubernetes Operators are quickly picking up traction in the developer community as a great way of managing complex applications on Kubernetes.